A potential financial relationship between investigators and pharmaceutical manufacturers has been associated with an increased likelihood of reporting favorable conclusions about a sponsor's proprietary agent in pharmacoeconomic studies. The purpose of this study is to investigate whether there is an association between financial relationships and outcome in economic analyses of new targeted therapies in oncology.
We searched PubMed (last update June 2011) for economic analyses of targeted therapies (including monoclonal antibodies, tyrosine-kinase inhibitors, and mammalian target of rapamycin inhibitors) in oncology. The trials were qualitatively rated regarding the cost assessment as favorable, neutral, or unfavorable on the basis of prespecified criteria.
Overall, 81 eligible studies were identified. Economic analyses that were funded by pharmaceutical companies were more likely to report favorable qualitative cost estimates (28 [82%] of 34 v 21 [45%] of 47; P = .003). The presence of an author affiliated with manufacturer was not associated with study outcome. Furthermore, if only studies including a conflict of interest statement were included (66 of 81), studies that reported any financial relationship with manufacturers (author affiliation and/or funding and/or other financial relationship) were more likely to report favorable results of targeted therapies compared with studies without financial relationship (32 [71%] of 45 v nine [43%] of 21; P = .025).
Our study reveals a potential threat for industry-related bias in economic analyses of targeted therapies in oncology in favor of analyses with financial relationships between authors and manufacturers. A more balanced funding of economic analyses from other sources may allow greater confidence in the interpretation of their results.